mm: set pte writable while pte_soft_dirty() is true in do_swap_page()

This patch leverages the new pte_needs_soft_dirty_wp() helper to optimize
a scenario where softdirty is enabled, but the softdirty flag has already
been set in do_swap_page().  In this situation, we can use pte_mkwrite
instead of applying write-protection since we don't depend on write
faults.

Link: https://lkml.kernel.org/r/20240607211358.4660-3-21cnbao@gmail.com
Signed-off-by: Barry Song <v-songbaohua@oppo.com>
Suggested-by: David Hildenbrand <david@redhat.com>
Acked-by: David Hildenbrand <david@redhat.com>
Cc: Chris Li <chrisl@kernel.org>
Cc: Kairui Song <kasong@tencent.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Ryan Roberts <ryan.roberts@arm.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
This commit is contained in:
Barry Song 2024-06-08 09:13:58 +12:00 committed by Andrew Morton
parent f38ee28519
commit 20dfa5b7ad

View File

@ -4324,7 +4324,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
if (!folio_test_ksm(folio) &&
(exclusive || folio_ref_count(folio) == 1)) {
if ((vma->vm_flags & VM_WRITE) && !userfaultfd_pte_wp(vma, pte) &&
!vma_soft_dirty_enabled(vma)) {
!pte_needs_soft_dirty_wp(vma, pte)) {
pte = pte_mkwrite(pte, vma);
if (vmf->flags & FAULT_FLAG_WRITE) {
pte = pte_mkdirty(pte);